| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, Arizona 85013
Submitted 20 December 2002; accepted in final form 23 April 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
;
Mountcastle 1995 for review).
Visual information and visuo-motor coordination play an important role in this
control. Despite the considerable amount of data on the activity of neurons in
the parietal cortex during different motor tasks, the neuronal mechanisms by
which visual information affects motor commands, as well as the principles of
visuo-motor coordination are not well understood. The goal of this study was
to characterize the activity of neurons in parietal area 5 during locomotion
under conditions that require visuo-motor coordination.
Area 5 is a high-order somato-sensory area, and it receives inputs from
various cortical and subcortical visual centers (Avendano et al.
1985,
1988;
Hendry et al. 1979;
Hyvarinen 1982;
Niimi et al. 1983;
Symonds et al. 1981). Area 5
provides outputs to the motor cortex, to other cortical and subcortical motor
centers, and to the spinal cord (Asanuma
1981; Babb et al.
1984; Ghosh 1997;
Groos et al. 1978;
Hyvarinen 1982;
Kakei et al. 1996;
Murray and Coulter 1981;
Wiesendanger 1981;
Wiesendanger et al. 1979;
Yumiya and Ghez 1984). Lesions
in area 5 cause difficulties in reaching to visual targets (Fabre and Buser
1980,
1981;
Rondot et al. 1977). The
activity of neurons in area 5 is correlated with reaching movements
(Bioulac et al. 1999;
Burbaud et al. 1985;
Ferraina and Bianchi 1994;
Hyvarinen 1982; Kalaska et al.
1983,
1990;
Khitrova-Orlova et al. 1997;
Mountcastle et al. 1975;
Sakata et al. 1995;
Scott et al. 1997;
Seal et al. 1982).
It was suggested that neuronal control of natural locomotion with precise
stepping uses the same neuronal machinery that is employed during reaching
movements (Georgopoulos and Grillner
1989). To walk in natural environments, a subject has to watch the
steps because the support surface is usually uneven. Visual information about
the surface structure allows the subject to perform necessary adjustments of
stepping. Thus natural locomotion involves two essentially different neural
mechanisms. First, basic rhythmical pattern of muscle contractions for each
limb is generated by the spinal cord (Forssberg et al.
1980a,b;
Grillner and Zangger 1979;
Sherrington 1906,
1910; for review, see
Orlovsky at al. 1999). Second,
the adaptations of this basic pattern to the features of natural supporting
surfaces are accomplished by supra-spinal mechanisms. It was shown, for
example, that the activity of the motor cortex changes substantially when cats
overstep obstacles or place paws precisely on cross-pieces of a horizontal
ladder during locomotion (Armstrong and
Drew 1984; Beloozerova and
Sirota 1993; Drew
1993; Widajewicz et al.
1994).
In the present study, we recorded the activity of neurons in area 5 in cats during three locomotion tasks: walking on a flat surface—a task with no visuo-motor coordination required (flat locomotion), walking on a horizontal ladder or overstepping a series of barriers—a task requiring visuo-motor coordination for accurate foot placement on a surface heterogeneous along the direction of progression (ladder and barriers locomotion), and walking along a narrow pathway—a task requiring visuomotor coordination for accurate foot placement on a surface homogeneous along the direction of progression (narrow locomotion). Comparison of neuronal discharges during these locomotion tasks provided evidence that neurons in area 5 of parietal cortex receive information about the activity of basic locomotion mechanisms and information about the heterogeneity of the supporting surface along the axis of progression. We describe here the modes of integration of these two types of information in the discharge patterns of area 5 neurons.
Some of the results have been reported in brief
(Beloozerova and Sirota
1992).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
;
Beloozerova et al. 2003) and
will be reported briefly here. All experiments were conducted with the
approval of the Barrow Neurological Institute Animal Care and Use
Committee. Experimental design
Positive reinforcement (food) was used to adapt cats to the experimental
situation and to induce locomotion (Pryor
1975). A box 2.5 m long and 0.5 m wide served as an experimental
chamber. A longitudinal wall divided the box into two corridors that cats
passed sequentially and repeatedly. In one of the corridors, the floor was
flat, and a horizontal ladder, barriers
(Fig. 1A), or
longitudinal dividers (Fig.
1B) were placed in the other. The cross-pieces of the
horizontal ladder were flat and 5 cm wide. The width of the cross-pieces was
chosen to exceed the cat's mean foot length that equals 3.5 mm, so that cats
had a full foot support on a crosspiece. Cross-pieces of the ladder were
spaced 25 cm apart, that is, at half of the mean step length observed during
flat locomotion (Beloozerova and Sirota
1993), and were elevated 8 cm above the ground. Barriers were 7 cm
high and 1 cm thick and were also placed 25 cm apart. The width of the narrow
walkway was 5 cm. The passage of a cat through the beginning or the end of
each corridor was monitored using infrared photodiodes.
|
Surgical procedures
Surgery was performed under isoflourane anesthesia using aseptic
procedures. The skin and fascia were removed from much of the dorsal surface
of the skull. At 10 points around the circumference of the head, stainless
steel screws were screwed into the skull and connected together with a wire;
the screw heads and the wire were then inserted into a plastic cast to form a
circular base. Later, awake cats were rigidly held by this base while
searching for neurons before a locomotion test. The base was also used later
for fixation of connectors, a miniature microdrive, and a protective and
electrically shielding cap. A portion of os parietale and the dura above
parietal area 5 over 
1.0 sq cm were removed. The area 5 region was
visually identified by the surface features. The aperture was then covered by
a 1-mm-thick plastic plate in which holes 0.36 mm in diameter had been drilled
and filled by sterile wax. The plate was fastened to the surrounding bone with
orthodontic resin (Densply Caulk).
Single-unit recording and neurons sampled
Sampling of neurons began several days after the surgery. The cat was
positioned on a table, and the base that was attached to its skull during the
surgery was fixed in the frame so that the head was immobilized while the body
was put in a comfortable position. Methods ensuring the humane treatment of
subjects during immobilization of the head have been described
(Beloozerova and Sirota 1993;
Beloozerova et al. 2003).
Neuronal activity was recorded extracellularly using either a conventional
tungsten-varnish-insulated (50–125 µm OD) microelectrode or a
quartz-insulated microelectrode with platinum-tungsten core (40 µm OD)
(Reitboeck 1983). The
electrode was advanced into cortical tissue through holes in the plastic plate
above area 5. A miniature manual single-axis micromanipulator, rigidly fixed
to the base, was used to lower the electrode. Positions of tracks from which
the activity of neurons was sampled, is schematically shown on
Fig. 1C. After
amplification and filtering (0.3- to 10-kHz band-pass; Power1401/Spike2
system, Cambridge Electronic Design, Cambridge, UK), unitary activity was
displayed on an oscilloscope and archived on a computer disk. A representative
record of a neuron discharge is shown in
Fig. 1D. To aid in the
identification of single neurons, the waveform analysis was employed to
discriminate and identify the spikes of the neuron using the Power1401/Spike2
system waveform-matching algorithm.
Receptive field testing
The somatic receptive fields were determined in resting animals by manual
application of stimuli to the skin and hair, by palpation of muscles and their
tendons, and by passive movements of joints. The responses to visual
stimulation were tested with the head restrained by switching the lights
ON-OFF by presenting stable and moving stimuli: two-dimensional
circles 2.5 cm in diameter, stripes 5 cm wide and 100 cm long, fields 50
x 50 cm, and complex three-dimensional stimuli: bars as those that were
used for cross-pieces on the ladder, toys, pieces of food, hands.
Two-dimensional stimuli were black and were presented against a white screen
that was positioned 50 cm in front of the animal. Stripes and fields were
oriented horizontal, vertical, or diagonal. Eight directions of planar
movements were tested: left-right, up-down, and four diagonal directions. In
addition, while an object was inside the receptive field of a neuron, the
screen was moved back and forward in the range of 25–75 cm in front of
the animal. Complex three-dimensional stimuli were presented against the
natural laboratory background and were moved in different directions in the
frontal plane at the distance of 50 cm in front of the animal and also
toward the animal and away from it. Speed of all stimuli movements was in the
range of 0.5–1.0 m/s, which is in the range of the speed of locomotion
that cats used inside the experimental chamber.
Processing of neuronal activity
The position and size of receptive fields was determined by listening to
the audio monitor and measuring the entire area from which action potentials
could be elicited. A directional selectivity was accessed by comparing the
number of spikes elicited by stimulation in different directions. During
locomotion, the duration of swing and stance phases of each step of the
forelimb contralateral to the cortical area under study was recorded using an
electrical sensor (Beloozerova and Sirota
1993). To accomplish this, a thin rubber sack was placed on the
cat's forelimb, a thin metal electrode was attached to the external surface of
the sack under the foot, and a voltage of 2–5 mV was applied to the
electrode. The floor of the experimental box was covered by a cloth moistened
with 0.1% sodium chloride to make the surface electrically conductive. A wire,
sewn into the cloth along the length of the box, was connected to the common
ground. The fall of voltage resulting from contact of the foot with the floor
was recorded. The start and finish of electrical contact was taken as the
start and finish of the stance phase of a step and, in selected experiments,
was verified using simultaneous video recording. The onset of the swing phase
was taken as the beginning of the step cycle. The duration of each step cycle
was divided into 10 equal bins, and the number of spikes in each bin was
counted. The discharge frequency in a bin was derived according to the method
of Udo et al. (1982) that
averages the instantaneous frequency of inter-spike intervals that fall within
the bin and also accounts for those intervals that overlap with bin's
beginning and end. The frequency histograms were smoothed using moving filter
with a span of three bins. The following parameters were calculated for each
neuron: the mean discharge frequency; the coefficient of modulation
(M) defined as
 |
Statistical procedures
Parametric tests were used when possible for comparisons between groups.
Unless noted otherwise, for all mean values, the SE is given. The discharge
frequency and modulation of neurons during different tasks was compared using
paired samples t-test. When data were categorical, nonparametric
2 or Mann-Whitney U tests were used.
Histological procedures
At the termination of the experiment, cats were deeply anesthetized with pentobarbital sodium and several reference electrolytic lesions were made in the area of recording. Cats were perfused with isotonic saline followed by a 10% Formalin solution. Frozen sections of 50-µm thickness were cut in the recording region. The tissue was stained for Nissl substance with cresyl violet. Positions of tracks in the cortex were estimated in relation to the reference lesions.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
We evaluated the involvement of vision in performance of locomotion tasks that we used for our experiments by testing the ability of cats to do these tasks in the complete darkness. Figure 2A shows the stepping pattern of a cat during walking along the horizontal ladder (ladder locomotion) and then during walking on the flat surface (flat locomotion) in an illuminated room. During the passage, the cat walked faster on the flat surface than on the ladder; however, the stepping patterns of each walk were rather regular. In the second trial (Fig. 2B), the lights were turned off when the cat entered the corridor with flat surface. On the flat surface, the cat stepped regularly in the dark and successfully reached the feeding rack. Thus vision was not necessary for flat locomotion. In contrast, when the lights were turned off while the cat was on the ladder in the third trial (Fig. 2C), locomotion was arrested: the cat did not place the foot on a support until after the lights were back on. Thus ladder locomotion required visual control of stepping. A cessation of locomotion was also observed in the fourth trial when the lights were turned off while the cat was in the corridor with barriers (Fig. 2D). Barriers locomotion did also require visual control of stepping. Walking along a narrow pathway (narrow locomotion) was possible, although irregular and greatly slowed in the dark (Fig. 2D). Thus narrow locomotion was dependent on visual control, but a partial compensation for the loss of vision was possible, probably by the use of somatosensory cues.
|
Activity of neurons during locomotion on the flat surface
The activity of 137 neurons was recorded during locomotion on the flat surface (flat locomotion). The mean speed of flat locomotion across the cats was 0.65 ± 0.14 (SD) m/s, and the step coefficient (the portion of swing phase in the step cycle) was 42 ± 1%. On transition from standing to locomotion, the discharge rate of many neurons increased, so the mean discharge rate for the whole population during locomotion was higher than that during standing (21.6 ± 1.7 vs. 16.3 ± 1.2 imp/s; P < 0.01, paired-samples t-test). During flat locomotion, the activity of 66% of all recorded neurons was modulated in the stepping rhythm; that is, the activity was higher in one phase of the step and lower in another phase (Fig. 3, A and B). The coefficient of modulation (M) was on average 58 ± 2%. Three patterns of modulation were observed. Among all recorded neurons, 27% had one peak of activity during step cycle (as the neuron in Fig. 3A), 19% had one trough (as the neuron in Fig. 3B), and 20% had two peaks (as the neuron in Fig. 3C). The neurons with one peak and one trough will be considered together in this paper and will be referred to collectively to as "one-peak" neurons (Fig. 3D). The bursts of activity of different neurons were distributed slightly unevenly over the step cycle. The bursts of some more neurons started around the beginning of swing and were well aligned to it (Fig. 3E). The mean duration of the bursts in this group was 30 ± 1% of the cycle. The bursts of somewhat smaller fraction of neurons started later in the cycle and were distributed with a more variable onset throughout the late swing and stance phases (Fig. 3E). In contract to the bursts of neurons from the first group, the duration of bursts in these neurons on average was longer (55 ± 3% of the cycle; P < 0.001, U test). The histogram of the mean frequencies of all step-related neurons had a weak peak during the swing phase (Fig. 3F).
|
Activity of neurons during locomotion on the ladder
On the horizontal ladder the surface was significantly heterogeneous along the direction of progression. During walking along the ladder (ladder locomotion), each limb overstepped two gaps and every other cross-piece of the ladder. The foot-prints of the two forelimbs are schematically shown in Fig. 1A as open circles on the cross-pieces of the ladder. The mean speed of ladder locomotion was 0.58 ± 0.07 (SD) m/s, and the step coefficient was 39 ± 1%.
All 137 neurons that were studied during flat locomotion were also studied during ladder locomotion. Three typical examples of the activity of neurons during flat and ladder locomotion are presented in Fig. 4. The activity of the neuron shown in A was not step-related during flat locomotion but had a double-peak discharge pattern during ladder locomotion. The neuron shown in B had one peak of activity during flat locomotion and two peaks during ladder locomotion, with each of the peaks being higher than that during flat locomotion. The neuron shown in C had two peaks during flat locomotion, and this discharge pattern became even more pronounced during ladder locomotion.
|
The discharge rate of 55% of neurons increased on transition from flat to
ladder locomotion, and the mean discharge rate of the whole population was
higher during ladder locomotion as compared with flat locomotion (29.1
± 2.6 vs. 21.6 ± 1.7 imp/s; P < 0.01, U
test). The fraction of modulated neurons was larger during ladder than during
flat locomotion (93 vs. 66%; P < 0.01, 2 test).
The average M also increased (82 ± 2 vs. 58 ± 2%; P
< 0.001, U test), as did the proportion of the double-peak neurons
(84% vs. 20%; P < 0.001, 2 test; Figs.
3D and
5A). In the
double-peak neurons, one peak usually occurred in swing, and the other one in
stance phase (Fig.
5B). The two bursts of activity were on average of an
approximately similar duration, occupying 30 ± 1% of the step cycle
each. In contrast to flat locomotion, there were much more cells the first
burst of activity in which started around the beginning of swing and much less
of those whose first bursts started later in the cycle (Figs.
3E and
5B). In 68% of the
double-peak neurons, the activity peak in swing was higher than that in
stance. This was not due to a better alignment of the histogram to the
beginning of swing because when aligned to the beginning of stance, the
histograms of the same portion of neurons still had the highest peak in swing
phase. The predominantly double peak pattern of the discharge was reflected in
the histogram of the mean frequencies of all step-related neurons that had two
peaks: one in the swing and another one in the stance phase
(Fig. 5C). The peak in
the swing phase was the stronger one.
|
Changes in the discharge pattern of individual neurons on transition from flat to ladder locomotion
NEURONS UNMODULATED DURING FLAT LOCOMOTION, N = 47(34% OF THE TOTAL SAMPLE). The activity of 79% (37/47) of these neurons became modulated in the rhythm of stepping during ladder locomotion. Sixty-six percent of them had a double-peak pattern (as the cell in Fig. 4A), and 13%, a single-peak one.
NEURONS WITH ONE PEAK OF ACTIVITY PER CYCLE DURING FLAT LOCOMOTION, N = 63 (46% OF THE TOTAL SAMPLE). In 90% (57/63) of these neurons there appeared a second peak of activity during ladder locomotion (as in cell in Fig. 4B). In addition, in one half of these neurons (28/57), the first peak changed its position in the step cycle on transition from flat to ladder locomotion (not illustrated). The two peaks always were approximately one half the duration of the cycle away from each other.
NEURONS WITH TWO PEAKS OF ACTIVITY PER CYCLE DURING FLAT LOCOMOTION N = 27 (20% OF THE TOTAL SAMPLE). All these neurons retained their double-peak discharge pattern during ladder locomotion as did the cell in Fig. 4C. The phase positions of the peaks in all but one of these cells remained unchanged.
Activity of neurons during stepping over a series of barriers
Although the width of the cross-pieces of the ladder exceeded the length of
cat's foot by 1.5 cm, we did an experiment to find out whether the
differences in activity of the neurons during flat and ladder locomotion were
due to the different requirements for the body stabilization during these two
locomotion tasks. We presented cats with a walkway partitioned with a series
of barriers. With the barriers, the surface of the corridor was significantly
heterogeneous along the direction of progression, as it was on the ladder;
however, the supporting surface was very wide (25 sq cm), thus the requirement
for the body stabilization was similar to that during flat locomotion. During
walking through a series of barriers (barriers locomotion), each limb
overstepped two successive barriers and every other gap between them
(Fig. 1A). The mean
speed of barriers locomotion was 0.65 ± 0.15 (SD) m/s, and the step
coefficient was 44 ± 1%.
Of 137 neurons recorded during flat and ladder locomotion, the activity of 47 neurons was also recorded during barriers locomotion. Figure 6, A–D, shows two typical examples of the activity of neurons during flat and barriers locomotion. The activity of the first neuron (A and B) was not step-related during flat locomotion (A) but had a clear-cut modulation with a two-peak pattern during barriers locomotion (B). The activity of the second neuron (C and D) had one peak per cycle during flat locomotion (C), and two peaks during barriers locomotion, each of which being higher than that during flat locomotion (D).
|
On transition from flat to barriers locomotion, the discharge rate of a
large number of neurons (43%) increased as it did on transition to ladder
locomotion, although in this case the mean activity increase for the whole
population did not reach the level of statistical significance (24.4 ±
3.0 imp/s as compared with 21.6 ± 1.7 imp/s during flat locomotion).
The fraction of modulated neurons was larger during barriers than during flat
locomotion (87 vs. 66%; P < 0.01, 2 test). The
average M was also larger (83 ± 2 vs. 58 ± 2%; P <
0.001, U test), as was the proportion of the double-peak neurons (70
vs. 20%; P < 0.001, 2 test; Figs.
6E and
3D). All the preceding
parameters were similar to those found during ladder locomotion. Although the
beginnings of the activity bursts were more variable during barriers than
during ladder locomotion, the predominantly double peak discharge pattern
during barriers locomotion was also reflected in the histogram of the mean
frequencies of the step-related neurons
(Fig. 6G). That
histogram had two peaks: one in the swing and another one in the stance
phase.
Activity of neurons during locomotion along the narrow walkway
In contrast to ladder and barriers locomotion, during walking along the narrow walkway (narrow locomotion), the surface on which cats stepped was homogeneous along the direction of progression. However, as during ladder and barriers locomotion, the feet had to be placed accurately on the narrow supporting surface. Thus using narrow locomotion task we could further test if the activity of area 5 neurons during locomotion depended on the precision of stepping or did it depend on irregularities in the visual scene. The mean speed of narrow locomotion was 0.57 ± 0.08 (SD) m/s, and the step coefficient was 43 ± 1%.
Of 137 neurons recorded during flat and ladder locomotion, the activity of 45 neurons was also recorded during narrow locomotion. Figure 7, A–C, gives a typical example of the activity of a neuron in 40 steps during each flat, narrow, and ladder locomotion. The discharge frequency of the neuron during flat locomotion (A) was low, but the activity did show a step-related modulation (the cell discharged more spikes at the end of stance phase than in any other portion of the step cycle). The discharge frequency of the neuron during narrow locomotion (B) was also low, and the step-related modulation was absent. When tested during ladder locomotion (C), however, the neuron demonstrated much higher activity and a clear-cut modulation that had two peaks: one approximately coinciding with the peak observed during flat locomotion and a newly formed one in early stance.
|
During narrow locomotion, the mean discharge rate of the neurons was similar to that during flat locomotion (25.0 ± 3.2 as compared with 21.6 ± 1.7 imp/s). The activity of 64% of neurons was modulated in the locomotion rhythm: 47% of them having single-peak pattern and 17% the double-peak pattern (proportions that were close to those found during flat locomotion, Figs. 7D and 3D). The average M was also similar to that during flat locomotion (57 ± 3 compared to 58 ± 2%). There were slightly more cells that had bursts in swing phase of the contralateral forelimb (Fig. 7E). The histogram of the mean frequencies of the step-related neurons had one peak in the swing phase (Fig. 7E). The bursts and the mean frequency distributions both were similar to those observed during flat locomotion (Fig. 3, E and F).
Activity of neurons with somato-sensory receptive fields
Somato-sensory receptive fields were found in 22% (14/64) of the neurons tested. The different receptive fields are summarized in Table 1.
|
During flat locomotion, the activity of only seven (50%) neurons with somatosensory receptive fields was step-related (Table 1), and there were five single-peak and two double-peak neurons. The mean frequency (19.0 ± 4.9 imp/s) and the average M (59 ± 5%) of these neurons were similar to those of the nonsomatosensory responsive ones (20.3 ± 1.7 imp/s and 58 ± 2%, respectively). In six neurons, a peak of activity occurred during swing phase of the contralateral forelimb. The activity of more than half of the neurons with receptive fields on the forelimb as well as both neurons with receptive fields on the neck was not step-related (Table 1).
During ladder locomotion, the activity of 12 (86%) neurons with somatosensory receptive fields was step-related (Table 1), and there were three single-peak and nine double-peak neurons. The mean activity (23.9 ± 4.8 imp/s) and the average M (87 ± 4%) of these neurons were again similar to those of the nonsomatosensory responsive ones (26.8 ± 2.5 imp/s and 82 ± 2%, respectively). In nine neurons, a peak of activity occurred during swing phase of the contralateral forelimb, and if there were two peaks, the one in the swing phase was the strongest one. The activity of one neuron with receptive field on the forelimb was still not step-related (Table 1).
Activity of neurons with visual receptive fields
Visual receptive fields were found in 20% (17/83) of the neurons tested. In all of these neurons, responses to lights ON-OFF and to stationary stimuli were weak or absent, however, all of the neurons responded well to moving stimuli. The receptive fields of all cells were excitatory and were located in the lower portion of the visual field in front of the animal. Except for three, all neurons responded to simple stimuli (black circles, stripes, or fields) moving in the frontal plane. Most neurons showed little or no directional selectivity in the frontal plane; however, all of them discharged more vigorously in response to approaching stimuli as compared with receding ones. All neurons responded to complex stimuli (bars, hands, toys, or food) with the approaching direction of their movement being the strongest stimulus.
At rest the activity of the visually responsive neurons was similar to that of the nonvisually responsive ones. With the start of locomotion, however, the activity of the visually responsive neurons decreased and during flat locomotion was lower than that of the nonvisually responsive ones (8.5 ± 2.0 vs. 21.3 ± 1.7 imp/s; P < 0.0001, U test). The activity of 60% of the visually responsive neurons was related to the locomotion cycle with an average M of 59 ± 6%, a value similar to that of the nonvisually responsive neurons (59 ± 3%). All of the visually responsive neurons had one peak of activity per step cycle, and in all of them, that peak occurred in the swing phase of the contralateral forelimb.
During ladder locomotion, the visually responsive neurons retained their low activity. The activity of 80% of them was modulated in the locomotion cycle; however, the average M was lower than that of the nonvisually responsive neurons (78 ± 4 vs. 82 ± 2%; P = 0.05, U test). Like the nonvisually responsive neurons, the great majority of the visually responsive neurons had two peaks of activity per step cycle. The phase distribution of the peaks of activity of the neurons with visual receptive fields was similar to that of the nonvisually responsive neurons.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
The first finding of the present study is that the activity of many neurons in parietal area 5 was step-related during locomotion on the flat surface, when visual control of stepping was not necessary. It can thus be concluded that area 5 receives movement-related information not only during the "high-order" arm reaching movements, but also during the "lower-order" spinally organized locomotion.
Is the step-related activity of area 5 neurons during flat locomotion
driven by somatosensory stimulation or is it driven by inputs from other motor
centers? At least three facts suggest that the activation of somatosensory
receptors does not importantly contribute to modulation of activity of neurons
in area 5 during locomotion. First, most of the neurons with the step-related
modulation of activity had no somatosensory receptive fields, at least those
that we could reveal with our testing technique in resting condition. Second,
the activity of four of seven neurons with somato-sensory receptive fields on
the forelimb was not step-related (Table
1). Third, an extra flexion of the limbs and thus an additional
activation of somatosensory receptors during overstepping series of barriers
as compared with that during ladder locomotion had little effect on the
activity of the neurons as they were rather similarly active and modulated
during both locomotion tasks (Figs.
6 and
5). Therefore it seems probable
that much like during arm reaching movements
(Bioulac et al. 1999;
Mountcastle et al. 1975; Seal
et al. 1982,
1983), activity of area 5
during flat locomotion to a significant extent depends on the status of other
motor-related brain centers and not on the status of peripheral receptors. The
similarity between phase distribution of the activity of neurons in area 5
(Fig. 3E) and spinal
interneurons (Orlovsky and Feldman
1972; Baev et al.
1979) suggests that spinal locomotion generator could be a source
of such internal information and thus be primarily responsible for modulation
of the activity of 66% of the neurons in area 5.
Activity of neurons during ladder locomotion—the roles of surface heterogeneity, precision of stepping, body stabilization, and direct visual stimulation
Visuo-motor coordination and, therefore supraspinal control of locomotion was required during ladder locomotion. The second finding of the present study was that the activity of neurons in area 5 was dramatically different during ladder locomotion as compared with flat locomotion. That difference was found in all of the measured parameters of the activity of most of the neurons: in the discharge rates, the coefficients of modulation, and the discharge patterns. Such pronounced differences strongly suggest that area 5 participates in visuomotor coordination during ladder locomotion. The question arises as to with what aspects of the coordination area 5 is concerned.
Unlike the flat surface, the horizontal ladder required precise placing of
feet on the support. From our previous work in the motor cortex, we know that
the requirement for precise feet placing may cause a dramatic change in the
activity of neurons (Beloozerova and
Sirota 1993). However, for area 5, we have earlier found that this
cortical area is not involved in the control of accuracy of stepping
(Beloozerova and Sirota 1992).
Indeed, in a test that involved overstepping a series of barriers spaced
progressively more closely to each other, thus gradually imposing more strict
requirements for precision of stepping, the neurons of area 5 failed to
correspondingly increase their discharges.
The potentially enhanced requirements for the body stabilization during ladder locomotion comparing to those during flat locomotion were also not responsible for the difference in the neuronal activity during the two locomotion tasks. In the control barriers locomotion test, the body stabilization requirements were close to those in the flat locomotion test; however, the discharges of the neurons were very different in these two tasks (Figs. 6 and 3); in contrast, they closely resembled the ones found during ladder locomotion (Figs. 6 and 5). The body stabilization was clearly compromised during narrow locomotion due to the unusual placing of right and left feet along a singe line; however, the neuronal discharges during narrow locomotion were very similar to those during flat locomotion.
We hypothesized that during ladder locomotion, the activity of area 5 is primarily concerned with the visual heterogeneity of the supporting surface rather than with the motor aspects of locomotion. To test this hypothesis, we have constructed a locomotion task that required precise placing of feet on the support but in a pathway that was visually homogeneous along the direction of progression—the task of narrow locomotion. It was found that during narrow locomotion the activity of neurons was similar to that during flat locomotion (Figs. 3 and 6). Therefore we concluded that the visual heterogeneity of the surface along the direction of progression was the main factor that modulated the activity of neurons during ladder locomotion in addition to their motor related modulation. The substantially different phase distributions of neuronal activity during ladder and flat locomotion (Figs. 5B and 3E) also suggest the very different driving forces for the activity modulation during these two tasks.
On the other hand, we have to conclude that the visual responsiveness of area 5 neurons, at least that that we could reveal with our testing technique in the resting conditions, was not responsible for modulation of neuronal activity during ladder locomotion. In our sample, there were a number of neurons that had visual receptive fields in the front of the animal. However, even the highly sensitive to incoming stimuli neurons, which seemed to be well situated to detect an approach of a transverse target during locomotion (a cross-piece of the ladder), demonstrated below the average levels of activity and step-related modulation during ladder locomotion. In contrast, neurons with no visual receptive fields at rest displayed a high involvement in visuo-motor coordination during locomotion. Although visual responsiveness of the neurons may be different during locomotion as compared with resting conditions, our findings allow to hypothesize that the visual peripheral information does not play much role in modulation of the activity of neurons of area 5 during locomotion, that it is the processed visual information that affects it in this situation.
Modes of integration of motor and visual information
The influence of visual information on the basic locomotion rhythm during ladder locomotion was exerted on the activity of the great majority of neurons. This influence was rather diverse, however.
For 37 neurons (27% of the total sample), the activity of which was not step-related during flat locomotion but was during ladder locomotion, visual information was the main modulator of their activity on the ladder (Fig. 8A). In addition, there were 28 neurons (20% of the total sample) in which visual information during ladder locomotion caused a change in the phase position of the activity peak observed during flat locomotion (Fig. 8B).
|
In contrast, visual information about the support surface during ladder locomotion had a relatively small effect on the discharge pattern of 27 neurons (20% of the total sample) the activity of which had two peaks per step cycle during flat locomotion (Fig. 8C). In the activity of these cells, the movement-related pattern was the dominant one. The role of visual information was to increase the activity levels and the efficacy of the step-related modulation (the coefficient of modulation, M).
There were also 29 neurons (21% of the total sample) in which the peak of discharge remained in place during transition from flat to ladder locomotion while the second peak emerged in the opposite phase of the step cycle (as in the cell shown in Fig. 4B). That group of neurons can be regarded as a group of "an equal representation," because in the activity of these neurons the motor-related modulation was present along with the visually driven modulation (Fig. 8D).
Because the activity of a majority of the neurons had two peaks per step
cycle during ladder locomotion, we believe that there were two visual messages
reflected in the activity of the neurons. These messages probably corresponded
to the results of the visual inspection of the two successive cross-pieces of
the ladder. Studies show that although humans and cats usually do not look
constantly at the landing targets when they walk with an accurate stepping
(Patla et al. 1996;
Patla 1997; Sherk and Fowler
2000,
2001), they do make
short-lasting visual inspection of them approximately two steps ahead
(Patla and Vickers 2003). When
the stepping task is challenging enough, however, saccades are almost always
made to the next footfall target (Hollands
and Marple-Horvat 2001). Obtained ahead of the time or immediately
before the step, information about positions of the targets on the surface
reaches area 5 and modifies activity of its neurons. The fact that the area 5
populational peak of activity during all locomotion tasks occurred during
swing phase of the contralateral forelimb suggests that area 5 was more
concerned with the position of the footfall target for that limb than with the
one for the ipsilateral limb. It remains unclear, however, if visual and motor
information are originally integrated on the neurons of area 5 or if area 5
receives this information already integrated from another motor center, for
example as an "efference copy" of a motor command.
How is the integrated visual and motor information used?
In the activity of area 5 neurons, visual information was integrated with locomotor rhythmicity by enhancing it and, in many instances, by adjusting its pattern. A question arises as to how this combined information is used for control of the limbs. To answer this question, recordings from identified efferent neurons of area 5 are required. At present we can only stress that it does not seems likely that parietal area 5 exerts its influence on the peripheral apparatus during locomotion via the motor cortex. There are at least three major features in the activity of area 5 and the motor cortex that argue against this route.
First, one of the major changes in the activity of area 5 on transition
from flat to ladder locomotion is the change in the pattern of activity of
most neurons from a single-peak to a double-peak pattern. As a result, the
majority of neurons in area 5 have two peaks of activity per step cycle during
ladder locomotion. During the same time, the predominant pattern of activity
of neurons in the motor cortex is the single-peak pattern
(Beloozerova and Sirota 1993;
Drew 1993;
Widajewicz et al. 1994). It
does not seem likely that the single-peak pattern of neurons in the motor
cortex is formed from the double-peak pattern of neurons in area 5.
Second, in a considerable number of neurons in area 5 the change in the
pattern of activity was accompanied by a change in its phase distribution. In
contrast, the characteristic feature of the activity of neurons in the motor
cortex during transition from flat to ladder locomotion is the rather stable
phase distribution of the activity in spite of a dramatic change in its
magnitude (Beloozerova and Sirota
1993). It seems unlikely that a stable phase distribution is
constructed out of a flexible one.
Finally, the activity of neurons in area 5 is not related to the precision
of stepping, while the activity of neurons in the motor cortex clearly is
(Beloozerova and Sirota
1993).
It seems most likely that area 5 acts independently of motor cortex
imposing its direct control on spinal cord and/or other subcortical motor
centers to which it projects (Asanuma
1981; Groos et al.
1978; Leichnetz et al.
1984; Murray and Coulter
1981; Wiesendanger
1981; Wiesendanger et al.
1979). For example, it was shown that the movements of the face
musculature evoked by microstimulation of area 5 are independent on the motor
cortex (Waters and Asanuma
1983). Thus visual information about heterogeneity of the support
surface in the direction of progression that is incorporated into step-related
discharges of neurons of area 5 promotes appropriate adjustments of stepping
on uneven support surfaces.
There were several hypotheses proposed for the role of area 5 in the
control of limb movements including the "sensory" hypothesis
(Hyvarinen 1982), the
"command" hypothesis
(Mountcastle et al. 1975), the
"planning action" hypothesis
(Andersen et al. 1998), the
"attention" hypothesis (Colby
and Goldberg 1999), the "decision" hypothesis
(Shadlen and Newsome 2001),
the "reference frames transformation" hypothesis
(Cohen and Andersen 2002), and
others. To a large extent, the diverse roles of area 5 may be complementary to
each other (Kalaska 1996) or
they may depend on the behavioral task. Our findings extend knowledge about
area 5 of the parietal cortex by showing for the first time how the neuronal
activity of this area reflects the process of integrating information about
the limb movements and the heterogeneity of the support surface during goal
directed stepping.
|
DISCLOSURES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests: I. N. Beloozerova, Neurobiology, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, 350 W. Thomas Rd., Phoenix, AZ 85013 (E-mail: ibelooz{at}chw.edu).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Andersen RA, Snyder LH, Bradley DC, and Xing J. Multimodal representation of space in the posterior parietal cortex and its use in planning movements. Annu Rev Neurosci 20: 303–330, 1997.[ISI][Medline]
Armstrong DM and Drew T. Discharge of pyramidal tract and other motor cortical neurons
during locomotion in the cat. J Physiol
346: 471–496,
1984.
Asanuma H. The pyramidal tract. In: Handbook of Physiology, Bethesda, MD: Am Physiol Soc, 1981, sect 1, vol. II, p. 703–733.
Avendano C, Rausell E, Perez-Aguilar D, and Isorna S. Organization of the association cortical afferent connections of area 5: a retrograde tracer study in the cat. J Comp Neurol 278: 1–33, 1988.[ISI][Medline]
Avendano C, Rausell E, and Reinoso-Suarez F. Thalamic projections to areas 5a and 5b of the parietal cortex in the cat: a retrograde horseradish peroxidase study. J Neurosci 5: 1446–1470, 1985.[Abstract]
Babb RS, Waters RS, and Asanuma H. Corticocortical connections to the motor cortex from the posterior parietal lobe (Areas 5a, 5b, 7) in the cat demonstrated by the retrograde axonal transport of horseradish peroxidase. Exp Brain Res 54: 476–484, 1984.[ISI][Medline]
Baev KV, Degtiarenko AM, Zavadskaia TV, and Kostiuk PG. Activity of interneurons of the lumbar region of the spinal cord during fictive locomotion of thalamic cats (in Russian). Neirofiziologiia 11: 329–338, 1979.[Medline]
Beloozerova IN and Sirota MG. Activity of neurons in area 5 of the cerebral cortex during locomotion in awake cats (Abstract). J Physiol 45: 291P, 1992.
Beloozerova IN and Sirota MG. The role of the motor cortex in the control of accuracy of
locomotor movements in the cat. J Physiol
461: 1–25,
1993.
Beloozerova IN,
Sirota MG, and Swadlow HA. Activity of different classes of neurons of the
motor cortex during locomotion. J Neurosci
23: 1087–1097,
2003.
Bioulac B, Seal J, Guehl D, Burbaud P, and Gross C. Reorganization of area 5 neuron activity in trained deafferented monkeys. Brain Res 835: 266–274, 1999.[ISI][Medline]
Burbaud P, Gross C, and Bioulac B. Peripheral inputs and early unit activity in area 5 of the monkey during a trained forelimb movement. Brain Res 337: 341–346, 1985.[ISI][Medline]
Cohen YE and Andersen RA. A common reference frame for movement plans in the posterior parietal cortex. Nat Rev Neurosci 3: 553–562, 2002.[ISI][Medline]
Colby CL and Goldberg ME. Space and attention in parietal cortex. Annu Rev Neurosci 22: 319–349, 1999.[ISI][Medline]
Drew T.
Motor cortical activity during voluntary gait modifications in the cat. I.
Cells related to the forelimbs. J Neurophysiol
70: 179–199,
1993.
Fabre M and Buser P. Structures involved in acquisition and performance of visually guided movements in the cat. Acta Neurobiol Exp 40: 95–116, 1980.[Medline]
Fabre M and Buser P. Effect of lesioning the anterior suprasylvian cortex on visuo-motor guidance performance in the cat. Exp Brain Res 41: 81–88, 1981.[ISI][Medline]
Ferraina S and Bianchi L. Posterior parietal cortex: functional properties of neurons in area 5 during an instructed-delay reaching task within different parts of space. Exp Brain Res 99: 175–178, 1994.[ISI][Medline]
Forssberg H, Grillner S, and Halberstma J. The locomotion of the low spinal cat. I. Coordination within a hindlimb. Acta Physiol Scand 168: 269–281, 1980a.
Forssberg H, Grillner S, Halberstma J, and Rossignol S. The locomotion of the low spinal cat. II. Interlimb coordination. Acta Physiol Scand 168: 283–295, 1980b.
Georgopoulos AP and Grillner S. Visuomotor coordination in reaching and locomotion.
Science 245:
1209–1210, 1989.
Ghosh S.
Comparison of the cortical connections of areas 4
and 4
in the
cat cerebral cortex. J Comp Neurol
388: 371–396,
1997.[ISI][Medline]
Grillner S and Zangger P. On the central generation of locomotion in the low spinal cat. Exp Brain Res 34: 241–261, 1979.[ISI][Medline]
Groos WP, Ewing LK, Carter CM, and Coulter JD. Organization of corticospinal neurons in the cat. Brain Res 143: 393–341, 1978.[ISI][Medline]
Hendry SHC, Jones EG, and Graham J. Thalamic relay nuclei for cerebellar and certain related fiber systems in the cat. J Comp Neurol 185: 679–714, 1979.[ISI][Medline]
Hollands MA and Marple-Horvat DE. Coordination of eye and leg movements during visually guided stepping. J Mot Behav 33: 205–216, 2001.[ISI][Medline]
Hyvarinen J. The Parietal Cortex of Monkey and Man. Studies of Brain Function. Berlin: Springer-Verlag, 1982, vol. 8.
Kakei S, Futami T, and Shinoda Y. Projection pattern of single corticocortical fibers from the parietal cortex to the motor cortex. Neuroreport 7: 2369–2372, 1996.[ISI][Medline]
Kalaska JF. Parietal cortex area 5 and visuomotor behavior. Can J Physiol Pharmacol 74: 483–498, 1996.[ISI][Medline]
Kalaska JF, Caminiti R, and Georgopoulos AP. Cortical mechanisms related to the direction of two-dimensional arm movements: relations in parietal area 5 and comparison with motor cortex. Exp Brain Res 51: 247–260, 1983.[ISI][Medline]
Kalaska JF, Cohen DA, Prud'homme M, and Hyde ML. Parietal area 5 neuronal activity encodes movement kinematics, not movement dynamics. Exp Brain Res 80: 351–364, 1990.[ISI][Medline]
Khitrova-Orlova TV, Sidyakin VG, Kulichenko AM, and Pavlenko VB. Pre-movement activity of neurons in the parietal associative cortex of the cat during different types of voluntary movement. Neurosci Behav Physiol 27: 82–86, 1997.[Medline]
Leichnetz GR, Smith DJ, and Spencer RF. Cortical projections to the paramedian tegmental and basilar pons in the monkey. J Comp Neurol 228: 388–408, 1984.[ISI][Medline]
Mountcastle VB. The parietal system and some higher brain
functions. Cereb Cortex 5:
377–390, 1995.
Mountcastle VB,
Lynch JC, Georgopoulos A, Sakata H, and Acuna C. Posterior parietal
association cortex of the monkey: command functions for operations within
extrapersonal space. J Neurophysiol
38: 871–908,
1975.
Murray EA and Coulter JD. Organization of corticospinal neurons in the monkey. J Comp Neurol 195: 339–365, 1981.[ISI][Medline]
Niimi K, Matsuoka H, Yamazaki Y, and Katayama T. Thalamic afferents to the anterior and middle suprasylvian gyri in the cat traced with horseradish peroxidase. J Hirnforsh 24: 173–187, 1983.
Orlovsky GN, Deliagina TG, and Grillner S. Neuronal Control of Locomotion. From Mollusc to Man, Oxford, UK: Oxford Univ. Press, 1999.
on the cross-pieces of the ladder and
in between the barriers schematically show placements of cat forelimbs.
B: placement of feet while walking along the narrow walkway.
C: positions of tracks from which the activity of neurons was sampled
is shown by dots on a schematic drawing of the brain (dorsal view).
D: a sample record of a neuron's discharge during flat and ladder
locomotion. Bottom: the swing (Sw, deflection up) and stance (St,
deflection down) phases of the step cycle of the forelimb contralateral to the
recording site in the cortex recorded by an electrical sensor.
, the time
periods when lights were turned off. A: a control passage in the
illuminated room. B: lights were turned off when the cat was walking
on the flat surface; there was no disturbance of locomotion. C:
lights were turned off when the cat was walking along the ladder; locomotion
was arrested and resumed only when the lights were back on. D: lights
were turned off when the cat was overstepping series of barriers; after 1 step
in the dark only locomotion was arrested and resumed only when the lights were
back on. E: lights were turned off when the cat was walking along the
narrow walkway; locomotion was slowed down.
) and stance (
